@inproceedings{louchheim-etal-2025-using,
    title = "Using Encipherment to Isolate Conditions for the Successful Fine-tuning of Massively Multilingual Translation Models",
    author = "Louchheim, Carter  and
      Sotnichenko, Denis  and
      Yamaguchi, Yukina  and
      Hopkins, Mark",
    editor = "Haddow, Barry  and
      Kocmi, Tom  and
      Koehn, Philipp  and
      Monz, Christof",
    booktitle = "Proceedings of the Tenth Conference on Machine Translation",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.wmt-1.14/",
    pages = "241--252",
    ISBN = "979-8-89176-341-8",
    abstract = "When fine-tuning massively multilingual translation models for low-resource languages, practitioners often include auxiliary languages to improve performance, but factors determining successful auxiliary language selection remain unclear. This paper investigates whether syntactic similarity or lexical overlap is more important for effective multilingual fine-tuning. We use encipherment to create controlled experimental conditions that disentangle these confounded factors, generating novel languages with identical syntax but no lexical overlap, and conversely, languages that preserve lexical overlap. Through extensive NLLB-200 fine-tuning experiments across Europarl and AmericasNLP datasets, we demonstrate that lexical overlap is the dominant factor. Syntactically identical auxiliary languages provide negligible benefits ( 1.0 ChrF), while languages with significant lexical overlap provide substantial improvements ( 5.0 ChrF), with effectiveness strongly correlated to KL-divergence between token distributions (r = -0.47, p .001). Our findings provide clear guidance: when selecting auxiliary languages for multilingual fine-tuning, prioritize lexical overlap over syntactic similarity."
}